US 11,733,386 B2
Methods and apparatuses for object detection in a scene based on lidar data and radar data of the scene
Koyel Banerjee, San Jose, CA (US); Sumanth Nirmal Gavarraju, San Jose, CA (US); and Mingkang He, Baltimore, MD (US)
Assigned to Bayerische Motoren Werke Aktiengesellschaft, Munich (DE)
Appl. No. 16/767,653
Filed by Bayerische Motoren Werke Aktiengesellschaft, Munich (DE)
PCT Filed Nov. 22, 2018, PCT No. PCT/EP2018/082145
§ 371(c)(1), (2) Date May 28, 2020,
PCT Pub. No. WO2019/154536, PCT Pub. Date Aug. 15, 2019.
Claims priority of application No. 18155945 (EP), filed on Feb. 9, 2018.
Prior Publication US 2020/0301013 A1, Sep. 24, 2020
Int. Cl. G01S 13/931 (2020.01); G01S 17/86 (2020.01); G01S 17/931 (2020.01); G01S 13/86 (2006.01)
CPC G01S 17/86 (2020.01) [G01S 13/865 (2013.01); G01S 13/931 (2013.01); G01S 17/931 (2020.01); G01S 2013/93271 (2020.01); G01S 2013/93275 (2020.01)] 11 Claims
OG exemplary drawing
 
1. A method of object detection in a scene based on lidar data and radar data of the scene, the method comprising:
transforming the lidar data and the radar data to a common coordinate system;
extracting different radar point clusters from the radar data;
extracting different lidar point clusters from the lidar data and associating each lidar point cluster with a respective target object;
estimating a target object's velocity based on the movement of the respective lidar point cluster between consecutive lidar images; and
comparing the estimated target object's velocity with velocity information of a corresponding radar point cluster to identify corresponding radar and lidar point clusters by comparing a velocity of the center of mass of a lidar point cluster with a velocity of the center of mass of a radar point cluster.